Moonshot AI

Complex problems, solved with ease AI Models is capability. Code, analyze, sheets and slides — simplify complex work with the latest flagship model.

Available models

Starting rate

452.2 RODI / 1M

Max context

Kimi K2.5moonshot-ai/kimi-k2.5

chatvisionreasoningcodingagentictoolslong-contextopen-sourceserverless

Kimi K2.5 is Moonshot AI's multimodal model for text and code generation, image understanding, tool use, and long-context reasoning.

Price

In452.2RODI/M~0.600 USD/M
Out2260.7RODI/M~3.00 USD/M

Context262KSpeedBalancedInput:textimageOutput:textModel details

Kimi K2.6moonshot-ai/kimi-k2.6

chatvisionmultimodalreasoningagenticcodinglong-contextopen-sourcetoolspreview

Kimi K2.6 via Microsoft Foundry on Azure (Moonshot AI). Latest Kimi frontier model with stronger long-horizon coding, agentic execution, and swarm-style task orchestration. Native multimodal (text + image), thinking and non-thinking modes, tool calling, and JSON mode. Direct from Azure — Global Standard deployment Kimi-K2.6.

Price

In715.9RODI/M~0.950 USD/M
Out3014.2RODI/M~4.00 USD/M
Cached120.6RODI/M~0.160 USD/M

Context262KSpeedBalancedInput:textimageOutput:textModel details

Kimi K3moonshot-ai/kimi-k3

chatvisionreasoningcodingagentictoolslong-contextopen-sourceserverless

Kimi K3 is Moonshot AI's open-weight model with native vision and a 1M-token context window for long-running coding and knowledge work. Text and image inputs, always-on reasoning, function calling, structured output, explicit prompt caching, and streaming.

Price

In2486.7RODI/M~3.30 USD/M
Out12433.4RODI/M~16.50 USD/M
Cached248.7RODI/M~0.330 USD/M

Context1MSpeedBalancedInput:textimageOutput:textModel details

Kimi K2 Thinkingmoonshot-ai/kimi-k2-thinking

chatreasoningcodingtoolslong-contextopen-sourceserverless

Kimi K2 Thinking is Moonshot AI's reasoning model for math, coding, and multi-step logic, with chain-of-thought enabled.

Price

In452.2RODI/M~0.600 USD/M
Out1883.9RODI/M~2.50 USD/M

Context262KSpeedDeepInput:textOutput:textModel details